Simulation and generalisation of agricultural resource use is problematical for three reasons. First, it is important how- and to which extent research results are generalised, because the choice of boundary conditions for research interferes with the conclusions. Second, the availability and accuracy of data interferes with the significance of the conclusions. Third, the combination of research from different disciplines is difficult, and requires communication across mindsets and institutional barriers. A framework of hierarchy and scale was defined with the aim to support and evaluate generalisation processes and the conveyance of information between sciences and between science and decision-makers. Within this framework, the four case studies of agricultural resource use in the form of fossil energy and nitrogen, respectively, were performed. The weaknesses of the generalisation procedures were related to the three points, stated above: the boundary definition, the data availability and the degree of interdisciplinarity. Further research in these questions is recommended.